More Pandora 2 crap (MicroSD Split)


Screen should be WXGA 16:10, not FHD


Audio may be good enough already


I think wowing people is largely about unusual details - he pandora was (and is) unusual because of the keyboard - game controls combo


on the topic of raided micro SD, a pair of microSD cards has an advantage compared to a single SD card; they won't die at exactly the same time.


I wouldn't mind:


full SDXC (external)


quad microSD RAID 5 (internal)


RAID and pocket computing don't go together, which is a good reason why they should!

I too wouldn't mind that!
 
Admittedly I'm looking out for self-interest here as much as other's: if Craig has his way and produces a $700 device with a bunch of excess features I don't feel add much to the device then I won't be getting it. The price would absolutely be justified in the capabilities it *could* provide, but not necessarily in the overall content.
That is one of the reasons I really like my twist on Binky's modular concept for stacks of two modules under the main unit - people could literally assemble what they want. It has disadvantages though (plastics quantity, net thickness, number of bits involved, etc...


The end result, though, is that the Pandora II needs to wow people as much as the Pandora did 4+ years ago when it was in development. To wow people is going to take some serious kit. I.e. it's going to take some $$$ to make a machine that does the title justice.


Quad core 2Ghz CPU - Tegra 3 or better?


2GB+ RAM


720p minimum, preferably 1080p screen (and video system to drive it)


Full on OpenGL graphics capable


Even better audio (if that is possible) from the MWeston studios.


Storage capacity at least as large as the Pandora - preferably larger, but that may be difficult.


Faster storage - 100MB/sec capable would not be out of line.


Stronger chassis/plastics


Updated/Improved WiFi (802.11n with 2.4 and 5Ghz)


Updated/Improved USB (3.0 both on the device and the host ports)


That's my list. Where do you think they should compromise?

Better not tie yourself to demands of full OpenCL requirements because chances are not good that any feasible asking price is going to let you accomplish that. Also massive lolz at starting off the list with "Tegra 3 or better" - way to aim high with an SoC that has been in devices for a year! Not that it can hit 2GHz anyway. Getting a good SoC is way more important than any single thing on your list.


Now here's my question - forget asking what needs to be compromised - how does this list drive a $699 price tag? When I say that I commend ED for talking about lowering the price it's because I don't think that a design with all this stuff actually must cost $699, even in relatively modest volumes (preferably a few tens of thousands, but < 10k if necessary). What they should do, IMO, is charge people a premium for the first batch of units (think in the hundreds, not thousands) to better fund development. Let the people like you who think nothing of the price pay the expense then let people later on get it cheaper once you've cleared the potential obstacles. By all costs do NOT do what they did this time, where people years later end up having to pay way more and only some people got it at the original price.
 
If you're going to use RAID5, then you best use at least 5 disks - that is where the optimum space hit for the parity disk starts to click.
Is there room for that many cards?


Maybe one SDXC slot on the mainboard, and a whole bunch of microSD on a daughterboard?
 
Given Ed's wonderful ability to post good sense and constructive news, I think it might help if there were an official post detailing the outline plans for the P2, addressing key points like:

  • Will it cost $750
  • Whats this magical hardware deal that requires android as a stock install?
  • Will it run real Linux out of the box
  • Are we restricted to microSD card slots?
  • How is the 8GB of ram going to be achieved?


Its early days yet, but the core customer base is being put off buying a product that hasn't really reached the drawing board - which is a crazy state of affairs :(
If microSD slots are used what is the max amount of slots we can have? I'd like to see 3 or even 4 slots. That way I could have 1 or 2 dedicated to music and videos the rest for games, os etc. I have A LOT of media and documents.

You know, if the P2 purposely used microSD to RAID them, I would probably be convinced that it was a good idea


Just as leaving space in the top corners of the iCP2 turns out to be for speakers, clearing the lid of the P2 for a huge screen


Another thought is that a premium console/palmtop should have a backpanel touchpad - PSvita style


Perhaps this explains the enthusiasm for android?

Not to get too far down this rabbit hole...


Two microSDXC slots = 1 full sized SD slot in fuction.


A four card microSDXC RAID10 would have roughly the same or less performance than a two card SDXC RAID1.


Now, if you're talking about a snap on module to the Binky modular design with my 'make both sides the same' addition - then you COULD make a single module with 12! microSD slots (3 dual stacks front and back) and run RAID 6 on those - and have a truly bitchin' palm sized server... I wonder if a surface to surface connector could handle the bandwidth to run something like that. Hmm...

Good infos. What are your thoughts on capacity versus number of slots vs cost for microSD compared against capacity versus number of slots vs cost for SD? For instance 4 32gb microSD =$84 (128gb total) vs 2 64gb SD= $200-$400. Then again I may be looking at it wrong and be missing certain other pros/cons. My hypothesis was that I could buy 4 cheap cards and upgrade overtime and still have more flexibility than the 2 SD card solution without adding much realestate. Also the microSD cards don't have to be right next to each other. At least one of them could be in the battery bay. Thoughts?

Sorry to all of you who have been through this conversation before. It was asked, so I will try to stay brief.


It isn't just about $/GB. You have greatly exaggerated the price for SD media.


Those 4 32GB microSD cards that you're getting for $84 are most likely Class 2 speed - and off brand at that. I was unable to find an example of your four for $84.


If you compare the lowest end of 64GB SDXC media, you can get to that kind of price or lower. Link for example only. 64GB SDXC for $33. I wouldn't buy it.


http://www.amazon.com/SCT-Digital-Ultimate-Extreme-Computer/dp/B008YL8GLW/ref=sr_1_24?s=pc&ie=UTF8&qid=1349128076&sr=1-24&keywords=64gb+sdxc


At that point you're, in theory, getting a much much faster card.


I prefer branded media. The story is the same there. If you look at any given level of technology (release dates approximate), SD media will always be double the capacity and double the speed of the largest and fastest microSD card available at the time. On the bottom end of the scale using old/deprecated cards you can get close in price GB/GB, but the SD card is always going to be faster and twice the capacity per card for roughly the same price per GB.


Since you're interested in 4x32gb capacity = 128GB, maybe you should look at something like this:


http://www.bhphotovideo.com/c/product/893909-REG/Kingston_SDX10V_128GB_128GB_SDXC_MEMORY_CARD.html


128GB for under $100 - and Class 10 storage at that. FAR short of the $200-$400 per 64GB that your post claimed. I realize it was, 'an exaggeration to prove a point,' but it was off by more than 400% over reality in price alone, not counting performance.
 
What they should do, IMO, is charge people a premium for the first batch of units (think in the hundreds, not thousands) to better fund development. Let the people like you who think nothing of the price pay the expense then let people later on get it cheaper once you've cleared the potential obstacles. By all costs do NOT do what they did this time, where people years later end up having to pay way more and only some people got it at the original price.
I'm fine with that. I would much rather have paid $500 for my first Pandora and gotten it in one year instead of 2.5. I was fine with paying the $700 for my 1Ghz Pandora even though I already had a Pandora, I could justify the upgraded version just on the time it saved me.


The initial people paying more - isn't that how a kick starter is actually supposed to work?


Come out with a $700 Kickstarter Pandora 2 limited edition of 1000 units - with the components and capabilities that would justify at least a $500+ price point. If I felt it met my needs, I'd likely get in line for it. Cripple it down to a $300 price point and I'm afraid that the device itself would have to be 'cheapened' far too much to get there.

The Pandora 2 forum is there for a reason, guys :p


Haha he's even pointing!
The Pandora 2 forum is there for a reason, guys :p
Maybe a mod could move these posts over into a new thread?

Agreed - we need a mod to create a thread in Pandora 2 SPAM and dump everything not pertaining at least in part to ED's plea into it.
 
Quad core 2Ghz CPU - Tegra 3 or better?


2GB+ RAM


720p minimum, preferably 1080p screen (and video system to drive it)


Full on OpenGL graphics capable


Even better audio (if that is possible) from the MWeston studios.


Storage capacity at least as large as the Pandora - preferably larger, but that may be difficult.


Faster storage - 100MB/sec capable would not be out of line.


Stronger chassis/plastics


Updated/Improved WiFi (802.11n with 2.4 and 5Ghz)


Updated/Improved USB (3.0 both on the device and the host ports)


That's my list. Where do you think they should compromise?

Quad core is not needed for me - dual core is enough, it's hard to get much gain from the 2 extra cores unless you're gonna use the device for number crunching. I could live with single core.


2GB of RAM would be nice indeed, 4GB would be wow, 8GB would be too much (I don't want to pay for memory I'm not gonna use).


I would like to have a 1280x800 or 1280x768 screen; I could live with 1280x720 if that has better availability. Full HD output over hdmi would be nice (I could live with only 720p and XGA output though), but a full HD screen is certainly not preferable to me unless it's cheaper and lower power than the alternatives.


OpenGL is not likely, and I don't think it's really needed - more and more things will be written with GL|ES in mind anyway.


I'd like to see the same 2 external SDXC slots, but faster. I don't think RAID over microSD is a practical thing. An internal microSD slot or a bigger NAND would be nice; an internal full SD slot would be wow.


I agree with the stronger case, upgraded wifi and USB.


What are the most expensive parts? Probably the SoC and the screen, right? I suggest not to use the bleeding edge ultimate specs for those, but rather the almost-as-good, readily available, tried and tested, more mature and much more bang for buck penultimate specs. Because in 6 months time (maybe even before the first units ship), the bleeding edge will be outdated again, and we will have paid $100 or $200 extra just to have been able to boast ultimate specs for half a year.
 
To come back to some stuff that's important for me:


The best solution of the SD Problem would be 1 Big and one micro external and a micro internal.


We should not leave away all the full sized slots but micro would be easyer for the board designer and nice for those who use micro SD in theire smartphones.


Also 4 shoulder buttons are needed.


Gamecube has 2 analog buttons and a third one.


PS2 also uses 4 Shoulder buttons.


Pandora 2 might be able to play PS2 and Gamecube so it's important to have good gaming controlls here.ö
 
What are the most expensive parts? Probably the SoC and the screen, right? I suggest not to use the bleeding edge ultimate specs for those, but rather the almost-as-good, readily available, tried and tested, more mature and much more bang for buck penultimate specs. Because in 6 months time (maybe even before the first units ship), the bleeding edge will be outdated again, and we will have paid $100 or $200 extra just to have been able to boast ultimate specs for half a year.

That is kind of the point I was trying to make. Parts hitting the market now are going to be out in other devices inside of 6 months and 2nd tier in 12 months - which means that they would be ideal for a device being designed today for sale in 12 months. Yes, they would still be expensive at the 12 month point - but they would be on their way down by then.


So the cost to produce the device would start off at a high price point to match it's recent technology - and go down as those components became cheaper. The key is to select components that will have 'legs' and be used in ours and other's devices for years to come.
 
Admittedly I'm looking out for self-interest here as much as other's: if Craig has his way and produces a $700 device with a bunch of excess features I don't feel add much to the device then I won't be getting it. The price would absolutely be justified in the capabilities it *could* provide, but not necessarily in the overall content.
That is one of the reasons I really like my twist on Binky's modular concept for stacks of two modules under the main unit - people could literally assemble what they want. It has disadvantages though (plastics quantity, net thickness, number of bits involved, etc...


The end result, though, is that the Pandora II needs to wow people as much as the Pandora did 4+ years ago when it was in development. To wow people is going to take some serious kit. I.e. it's going to take some $$$ to make a machine that does the title justice.


Quad core 2Ghz CPU - Tegra 3 or better?


2GB+ RAM


720p minimum, preferably 1080p screen (and video system to drive it)


Full on OpenGL graphics capable


Even better audio (if that is possible) from the MWeston studios.


Storage capacity at least as large as the Pandora - preferably larger, but that may be difficult.


Faster storage - 100MB/sec capable would not be out of line.


Stronger chassis/plastics


Updated/Improved WiFi (802.11n with 2.4 and 5Ghz)


Updated/Improved USB (3.0 both on the device and the host ports)


That's my list. Where do you think they should compromise?

This sounds a little to me like:


Alright, we made a great movie, its geeky and was catching people with its on charm. We should make a second movie, this time much better. More explosion, better special effects more awesome cast. Most of those sequels turn out shitty because they loose their charm.


Also, i do not want to find myself in the position of walking around and running into those situations where you start comparing your hardwares bang with some else's hardwares bang. I prefer the felling of supporting the right thing and this feeling get undermined badly by this more power craze.


I mentioned it once, it went though under in the conversation that was already ongoing.


This is a nice option for a modular CPU i think:


http://rhombus-tech.net/


Other community projects might rely on this one too which (I can imagine) will make it a cheaper option and help modularity. We would support them and other projects by just buying their module. This would move complexity away (at least thats how I imagine it again) and allows to focus on oder parts, like extension systems for other modules, more time to choose the right peripherals...


Otherwise, this is a very powerful module from samsung right now (Hopefully cheaper when ordered in bigger quantities... I still would prefer to take the former project into account):


http://www.hardkerne...=G134032695534#


I also found this Kickstarter project interesting: http://www.kickstart...ne?ref=category


I imagine someone could use such processor grids for many interesting things (I personally wonder if one could implement a software graphics card on it and how well that would work. Might solve graphics driver issues!? Emulators for you gamers out there?).


I hope to hold one day the Pandora evolution in my hands and not the awesome super fancy, 1000 Tera Hertz, USB5.0, and whatnot Pandora. Would probably be like a bad sequell. what looks good on paper often isn't in reality. The focus is imporatnt.


Based on what i read on the forums, this cold be improved, case (lid) design, keypad fine tunings for better writing speed and comfort; a CPU upgrade which is powerful enough to emulate some more PC games (see also the kickstarter above ;) ); more reliable wifi; USB1.1 support; second set of shoulder buttons; shoulder buttons could improve mechanically a little; .... . I personally would wish for display signals on the extension connector, and some other signals multiplexed (USART, SPI, I2C, USB, mybe even SATA...).


What should not change: The form and size; the basic keyboard layout; the sound quality (speakers and especially headphone jack); the nubs and other game controls quality; the display quality; the touchscreen; the stability of the case; the default OS; battery time; physical volume wheel; and for sure some more things.


My dream Pandora is exactly like the pandora right now. Quality never gets old and i am a realist. Finally, the community was very often mentioned as a reason for the awesomeness of the pandora. I would say keeping the price in mind, trying to improving the not so good sides and keeping the not so bad sides, then the nice community will live on around this nice device and maybe attract some more people with a stable concept :wub: .


PS: The pandora wowed me in 2010 when i first saw it and still does...
 
Last edited by a moderator:
^ agree with above.


The fastest hand-held yet marketing was a bad move in hindsight.


I think going with versatile is a good plan.


An ideal, If impractical Pandora might have:


GPS


DAB/DVB


Wifi


Bluetooth


HSPDA


wired ethernet


eSATA


USB host


USB slave


HDMI out


audio OUT


audio IN


mic IN


heaps of buttons


lots of indicator lights


a touchpad


a touchscreen


gaming controls a keyboard


A screen suitable for native desktop software (not too big, not too small)


stereo speakers


I2C


SDIO


lots of SD / microSD slots


built-in storage for the OS


a hardware expansion system


Imagine a big clunky laptop that shrunk in the wash, but miraculously still works :)
 
Pandora 2 might be able to play PS2 and Gamecube so it's important to have good gaming controlls here.ö
I don't think that will be possible. I couldn't even emulate a Nintendo DS full speed on my Core 2 Duo notebook (1,6 Ghz).


Look at the recommended PCSX2 hardware specs: Intel Core 2 Duo @ 3.2 GHz or better OR i3/i5/i7 @ 2,8 GHz or better OR AMD Phenom II @ 3,2 GHz or better GPU: 8800gt or better (for Direct3D10 support) RAM: 1GB on Linux/Windows XP, 2GB or more on Vista / Windows 7
 
Last edited by a moderator:
Given Ed's wonderful ability to post good sense and constructive news, I think it might help if there were an official post detailing the outline plans for the P2, addressing key points like:

  • Will it cost $750
  • Whats this magical hardware deal that requires android as a stock install?
  • Will it run real Linux out of the box
  • Are we restricted to microSD card slots?
  • How is the 8GB of ram going to be achieved?


Its early days yet, but the core customer base is being put off buying a product that hasn't really reached the drawing board - which is a crazy state of affairs :(
If microSD slots are used what is the max amount of slots we can have? I'd like to see 3 or even 4 slots. That way I could have 1 or 2 dedicated to music and videos the rest for games, os etc. I have A LOT of media and documents.

You know, if the P2 purposely used microSD to RAID them, I would probably be convinced that it was a good idea


Just as leaving space in the top corners of the iCP2 turns out to be for speakers, clearing the lid of the P2 for a huge screen


Another thought is that a premium console/palmtop should have a backpanel touchpad - PSvita style


Perhaps this explains the enthusiasm for android?

Not to get too far down this rabbit hole...


Two microSDXC slots = 1 full sized SD slot in fuction.


A four card microSDXC RAID10 would have roughly the same or less performance than a two card SDXC RAID1.


Now, if you're talking about a snap on module to the Binky modular design with my 'make both sides the same' addition - then you COULD make a single module with 12! microSD slots (3 dual stacks front and back) and run RAID 6 on those - and have a truly bitchin' palm sized server... I wonder if a surface to surface connector could handle the bandwidth to run something like that. Hmm...

Good infos. What are your thoughts on capacity versus number of slots vs cost for microSD compared against capacity versus number of slots vs cost for SD? For instance 4 32gb microSD =$84 (128gb total) vs 2 64gb SD= $200-$400. Then again I may be looking at it wrong and be missing certain other pros/cons. My hypothesis was that I could buy 4 cheap cards and upgrade overtime and still have more flexibility than the 2 SD card solution without adding much realestate. Also the microSD cards don't have to be right next to each other. At least one of them could be in the battery bay. Thoughts?

Sorry to all of you who have been through this conversation before. It was asked, so I will try to stay brief.


It isn't just about $/GB. You have greatly exaggerated the price for SD media.[Actually I used google results :/]


Those 4 32GB microSD cards that you're getting for $84 are most likely Class 2 speed - and off brand at that. I was unable to find an example of your four for $84. [good thing I provided examples of class 4 NAMEBRANDS then]


If you compare the lowest end of 64GB SDXC media, you can get to that kind of price or lower. Link for example only. 64GB SDXC for $33. I wouldn't buy it.


http://www.amazon.co...words=64gb sdxc


At that point you're, in theory, getting a much much faster card.


I prefer branded media. The story is the same there. If you look at any given level of technology (release dates approximate), SD media will always be double the capacity and double the speed of the largest and fastest microSD card available at the time. On the bottom end of the scale using old/deprecated cards you can get close in price GB/GB, but the SD card is always going to be faster and twice the capacity per card for roughly the same price per GB.


Since you're interested in 4x32gb capacity = 128GB, maybe you should look at something like this:


http://www.bhphotovi...EMORY_CARD.html


128GB for under $100 - and Class 10 storage at that. FAR short of the $200-$400 per 64GB that your post claimed.[my post claimed $200-400 for 128gb mister math fail] I realize it was, 'an exaggeration to prove a point,' but it was off by more than 400% over reality in price alone, not counting performance.[if you paid attention in school maybe your math wouldn't be so bad and you wouldn't jump to the wrong conclusions. My God you even quoted exactly what I said! "(128gb total) vs 2 64gb SD= $200-$400" :p ]
So my original reply got eaten during the move. I did a google cache to recover it. Necessary because unfair allegations were levied my way. :p


Hi this is what I based my post off. I don't know if the search return is the same in your location. 32gb Sd and 64 gb SD. The top result for the 64 gb had a sandisk class 10 I believe for $93 the other was a Lexar for $199 (yes both are class 10). Now the micro SD card price I used was this one which is a class 4. So I wasn't exaggerating to prove a point. That's not my style entirely. However I didn't bother to try to find a class 4 64gb SD card. My fault, I was just doing a quick google. So here is a search for class 4 64gb SD cards.


These are also Sandisc with a price range of $59.95 to $77.97 (so $120 to $156 for 128gb)[<-this is to help compare between Class 4 micro SD's @ $21 and Class 4 SD cards]. So in short my claim of $200-$400 was correct in regards to namebrand 64gb CLASS 10 cards.


In fact the was zero exaggeration in my post as the provided links illustrate. I do however, admit to not doing the due diligence to make sure I was comparing the correct classes against each other. Even when the correct classes of both card types are used to compare cost higher gb SD cards are more expensive than lower gb microSD cards generally speaking and dealing with name brand cards. If you find I'm still in error please feel free to correct me as you might have found a really good deal on cards.


However intuitively speaking I don't anticipate a card with double the capicity of the same class micro SD card being cheaper than two microSD cards half the capacity. Here's a Sandisc 32gb microSD for $12.53


If you've found namebrand class 4 or greater SD cards of high capacity cheaper than 4 lower capacity microSD cards of the same speed please let us know as we all are interested in a good deal!


Actually heres a selection of 64gb class 10 SD cards ranging in price from $48-$349.


And here's a selection of 32gb class 10 microSD cards ranging in price from $20!-$43(average around $25). Both SD and microSD cards priced are all name brands.
 
Last edited by a moderator:
OK. If you think microSD is better, please show me a 128GB class 10 write and 600X read card. That is 128GB capacity, 90MB/sec read and 45MB/sec write - all in one card.


I'll even hold my breath while you look - OK, that part was a lie.


The underlying fact is that you can not compare a stack of four 32GB class 4 microSD cards to a class 10 600X SDXC UHS-I card. It's apples to oranges - and 4 year old tech vs bleeding edge. The fact that you can get a 128GB SDXC card for roughly the same price as four old-technology 32GB microSD cards isn't exactly praise for microSD.


Compare tech level to tech level, capacity to capacity, price to price - full sized SD will always win.
 
What are you going on about? I set the goal post at 128gb between either 2 SD cards or 4 microSD cards. I never said microSD's were better, I said they were more cost effective! Further I fully compared APPLES TO APPLES as I provided four links, 2 each for SD and 2 each for micro that were for CLASS 10 CARDS IN ALL CASES.


Please review them. You should find that they support my assertion that 4 lower capacity cards are cheaper than 2 large capacity cards. It's simple economics. The larger the capacity the more cost that is associated with its manufacture. Why would you argue against that after I compared class 4 sd/micro's and then clas 10 sd/micros? In each case the micro was cheaper for reasons already discussed. If there were an option for 4 SD slots then my argument would be 4 lower capacity SD slots cost less than two large or even a single large capacity card.


I will bold the links so that you can see them easier and maybe you'll finally start talking about the presented facts and not what you imagine that you read.

However intuitively speaking I don't anticipate a card with double the capicity [<-class 10 SD cards]of the same class micro SD [<-also class 10 cards]card being cheaper than two microSD cards half the capacity. Here's a Sandisc 32gb microSD for $12.53


If you've found namebrand class 4 or greater SD cards of high capacity cheaper than 4 lower capacity microSD cards of the same speed please let us know as we all are interested in a good deal!


Actually heres a selection of 64gb class 10 SD cards [<--a selection of class 10 SD cards]ranging in price from $48-$349.


And here's a selection of 32gb class 10 microSD cards [<--a selection of class 10 microSD cards] ranging in price from $20!-$43(average around $25). Both SD and microSD cards priced are all name brands.

My whole point is that it is cheaper to buy lower capacity cards than one 128gb card that's only cheap if it's an off brand. You do like name brands right?


Edit: Also I would appreciate it if you refrain from accusing me of exagerating. My original response showed that without a doubt you had no idea what you were talking about. This time around try to stay on the same page. If even after this you still can't understand, let's take it to pm's.
 
Last edited by a moderator:
When I do the comparison using froogle, I get:


SanDisk Extreme HD Video Flash memory card - 128 GB SDXC UHS-I Memory Card - around £100


SanDisk Ultra Flash memory card - 64 GB microSDXC UHS-I Memory Card - a little under £40


I don't know the difference between "Extreme HD Video" and "Ultra" (these were the first results for "sandisk 128 SDXC card"), but the above suggests that you could save £20 by using a pair of microSD cards.
 
Back
Top